Galerie de prompts IA OpenZhenZhen · MiniMax H3 · video

Le double d'identité dans le cabinet aux murs de pierre

The Identity Double in the Stone-Walled Study

Le double d'identité dans le cabinet aux murs de pierre — découvrez cette création avec MiniMax H3 sur OpenZhenZhen, son prompt original et un exemple de génération à réutiliser.

Le double d'identité dans le cabinet aux murs de pierre — AI video creation
Le double d'identité dans le cabinet aux murs de pierre

Prompt

Prompt chinois

生成一段虚构的20秒电影感视频,画面比例为16:9。场景是夜晚昏暗的现代书房,墙面由大型石块砌成,配有木制家具、书桌和漆黑窗户,氛围严肃而私密,色彩克制,采用写实电影灯光。镜头1(0–5秒):固定俯拍机位、对称构图的局部极近景。一名白人老年高级政府官员快速敲击笔记本电脑。切至屏幕:虚构的身份分析界面显示两名男子的侧面与正面头像,开始进行面部识别和结构比对。他以平静严肃的语气说:“我找不到你,所以只能找一个人来冒充你。”镜头2(5–10秒):固定平视机位、居中构图的肩部以上近景。一名留着中等长度棕发、穿米色夹克的年轻白人特工出现,面部轮廓清晰,神情凝重,目光紧盯屏幕。官员在画外平静地说:“萨姆是完美人选。他曾经两三次冒充过你。”镜头3(10–15秒):固定平视机位、对称构图的笔记本屏幕特写。界面动态图形显示两张脸重叠、旋转比对,并分析抽象的音频波形,突出虚构的高科技身份模仿技术。显示系统提示文字“Sound correction(声音校正)”。镜头4(15–20秒):回到特工相同的居中平视肩部以上近景。他直视前方,嘴唇轻微移动,果断回答:“两次。”保持灯光、服装、场景和角色外观连续一致,使用准确剪辑、细微屏幕反光、自然面部动作和清晰对白节奏。不得出现真实人物、可识别目标、源视频或源音频。

Prompt anglais

Generate a fictional 20-second cinematic sequence in a 16:9 aspect ratio. Nighttime, dimly lit modern study with large stone-block walls, wooden furniture, a desk, and pitch-black windows; serious, private atmosphere, restrained color palette, realistic film lighting. Shot 1 (0–5s): Partial extreme close-up from a fixed overhead camera, symmetrical composition. An elderly white male senior government official rapidly types on a laptop. Cut to the laptop display: a fictional identity-analysis interface shows profile and frontal headshots of two men while facial recognition and structural comparison begin. He says calmly and seriously, “I can’t find you, so I can only find someone to pretend to be you.” Shot 2 (5–10s): Fixed eye-level, centered shoulders-up close-up of a young white male agent with medium-length brown hair and a beige jacket. His defined facial contours and grave expression remain fixed on the screen. The official speaks off-screen: “Sam is the perfect choice. He pretended to be you two or three times.” Shot 3 (10–15s): Fixed eye-level symmetrical close-up of the laptop. Interface graphics dynamically show two faces overlapping, rotating for comparison, and abstract audio waveform analysis, emphasizing fictional high-tech identity imitation. Display a system message reading “Sound correction.” Shot 4 (15–20s): Return to the agent in the same centered, eye-level shoulders-up close-up. He looks straight ahead, lips moving slightly, and decisively replies, “Twice.” Maintain continuity of lighting, wardrobe, set, and character appearance; use precise cuts, subtle screen reflections, natural facial motion, and clear dialogue timing. No real persons, identifiable targets, source footage, or source audio.

Thèmes et styles

Voir le modèle API

Découvrir plus de prompts image et vidéo · Générer des images · Générer des vidéos · Créer sur le canevas IA